#include <linux/config.h>
#include <linux/time.h>
+#include <linux/capability.h>
#include <linux/fs.h>
#include <linux/jbd.h>
#include <linux/ext3_fs.h>
struct buffer_head ** bh)
{
unsigned long group_desc;
- unsigned long desc;
- struct ext3_group_desc * gdp;
+ unsigned long offset;
+ struct ext3_group_desc * desc;
+ struct ext3_sb_info *sbi = EXT3_SB(sb);
- if (block_group >= EXT3_SB(sb)->s_groups_count) {
+ if (block_group >= sbi->s_groups_count) {
ext3_error (sb, "ext3_get_group_desc",
"block_group >= groups_count - "
"block_group = %d, groups_count = %lu",
- block_group, EXT3_SB(sb)->s_groups_count);
+ block_group, sbi->s_groups_count);
return NULL;
}
smp_rmb();
- group_desc = block_group / EXT3_DESC_PER_BLOCK(sb);
- desc = block_group % EXT3_DESC_PER_BLOCK(sb);
- if (!EXT3_SB(sb)->s_group_desc[group_desc]) {
+ group_desc = block_group >> EXT3_DESC_PER_BLOCK_BITS(sb);
+ offset = block_group & (EXT3_DESC_PER_BLOCK(sb) - 1);
+ if (!sbi->s_group_desc[group_desc]) {
ext3_error (sb, "ext3_get_group_desc",
"Group descriptor not loaded - "
"block_group = %d, group_desc = %lu, desc = %lu",
- block_group, group_desc, desc);
+ block_group, group_desc, offset);
return NULL;
}
- gdp = (struct ext3_group_desc *)
- EXT3_SB(sb)->s_group_desc[group_desc]->b_data;
+ desc = (struct ext3_group_desc *) sbi->s_group_desc[group_desc]->b_data;
if (bh)
- *bh = EXT3_SB(sb)->s_group_desc[group_desc];
- return gdp + desc;
+ *bh = sbi->s_group_desc[group_desc];
+ return desc + offset;
}
/*
* code.
*/
#if 0
-static void __rsv_window_dump(struct rb_root *root, int verbose,
+static void __rsv_window_dump(struct rb_root *root, int verbose,
const char *fn)
{
struct rb_node *n;
- struct reserve_window_node *rsv, *prev;
+ struct ext3_reserve_window_node *rsv, *prev;
int bad;
-
+
restart:
n = rb_first(root);
bad = 0;
prev = NULL;
-
+
printk("Block Allocation Reservation Windows Map (%s):\n", fn);
while (n) {
- rsv = list_entry(n, struct reserve_window_node, rsv_node);
+ rsv = list_entry(n, struct ext3_reserve_window_node, rsv_node);
if (verbose)
printk("reservation window 0x%p "
"start: %d, end: %d\n",
rsv, rsv->rsv_start, rsv->rsv_end);
if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) {
- printk("Bad reservation %p (start >= end)\n",
+ printk("Bad reservation %p (start >= end)\n",
rsv);
bad = 1;
}
#endif
static int
-goal_in_my_reservation(struct reserve_window *rsv, int goal,
+goal_in_my_reservation(struct ext3_reserve_window *rsv, int goal,
unsigned int group, struct super_block * sb)
{
unsigned long group_first_block, group_last_block;
return 1;
}
-/*
+/*
* Find the reserved window which includes the goal, or the previous one
* if the goal is not in any window.
* Returns NULL if there are no windows or if all windows start after the goal.
*/
-static struct reserve_window_node *search_reserve_window(struct rb_root *root,
- unsigned long goal)
+static struct ext3_reserve_window_node *
+search_reserve_window(struct rb_root *root, unsigned long goal)
{
struct rb_node *n = root->rb_node;
- struct reserve_window_node *rsv;
+ struct ext3_reserve_window_node *rsv;
if (!n)
return NULL;
-
- while (n)
- {
- rsv = rb_entry(n, struct reserve_window_node, rsv_node);
+
+ do {
+ rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node);
if (goal < rsv->rsv_start)
n = n->rb_left;
n = n->rb_right;
else
return rsv;
- }
- /*
+ } while (n);
+ /*
* We've fallen off the end of the tree: the goal wasn't inside
* any particular node. OK, the previous node must be to one
* side of the interval containing the goal. If it's the RHS,
- * we need to back up one.
+ * we need to back up one.
*/
if (rsv->rsv_start > goal) {
n = rb_prev(&rsv->rsv_node);
- rsv = rb_entry(n, struct reserve_window_node, rsv_node);
+ rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node);
}
- return rsv;
+ return rsv;
}
-void rsv_window_add(struct super_block *sb,
- struct reserve_window_node *rsv)
+void ext3_rsv_window_add(struct super_block *sb,
+ struct ext3_reserve_window_node *rsv)
{
struct rb_root *root = &EXT3_SB(sb)->s_rsv_window_root;
struct rb_node *node = &rsv->rsv_node;
struct rb_node ** p = &root->rb_node;
struct rb_node * parent = NULL;
- struct reserve_window_node *this;
-
+ struct ext3_reserve_window_node *this;
+
while (*p)
{
parent = *p;
- this = rb_entry(parent, struct reserve_window_node, rsv_node);
+ this = rb_entry(parent, struct ext3_reserve_window_node, rsv_node);
if (start < this->rsv_start)
p = &(*p)->rb_left;
}
static void rsv_window_remove(struct super_block *sb,
- struct reserve_window_node *rsv)
+ struct ext3_reserve_window_node *rsv)
{
rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
- atomic_set(&rsv->rsv_alloc_hit, 0);
+ rsv->rsv_alloc_hit = 0;
rb_erase(&rsv->rsv_node, &EXT3_SB(sb)->s_rsv_window_root);
}
-static inline int rsv_is_empty(struct reserve_window *rsv)
+static inline int rsv_is_empty(struct ext3_reserve_window *rsv)
{
/* a valid reservation end block could not be 0 */
return (rsv->_rsv_end == EXT3_RESERVE_WINDOW_NOT_ALLOCATED);
}
+void ext3_init_block_alloc_info(struct inode *inode)
+{
+ struct ext3_inode_info *ei = EXT3_I(inode);
+ struct ext3_block_alloc_info *block_i = ei->i_block_alloc_info;
+ struct super_block *sb = inode->i_sb;
+
+ block_i = kmalloc(sizeof(*block_i), GFP_NOFS);
+ if (block_i) {
+ struct ext3_reserve_window_node *rsv = &block_i->rsv_window_node;
+
+ rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
+ rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
+
+ /*
+ * if filesystem is mounted with NORESERVATION, the goal
+ * reservation window size is set to zero to indicate
+ * block reservation is off
+ */
+ if (!test_opt(sb, RESERVATION))
+ rsv->rsv_goal_size = 0;
+ else
+ rsv->rsv_goal_size = EXT3_DEFAULT_RESERVE_BLOCKS;
+ rsv->rsv_alloc_hit = 0;
+ block_i->last_alloc_logical_block = 0;
+ block_i->last_alloc_physical_block = 0;
+ }
+ ei->i_block_alloc_info = block_i;
+}
void ext3_discard_reservation(struct inode *inode)
{
struct ext3_inode_info *ei = EXT3_I(inode);
- struct reserve_window_node *rsv = &ei->i_rsv_window;
+ struct ext3_block_alloc_info *block_i = ei->i_block_alloc_info;
+ struct ext3_reserve_window_node *rsv;
spinlock_t *rsv_lock = &EXT3_SB(inode->i_sb)->s_rsv_window_lock;
+ if (!block_i)
+ return;
+
+ rsv = &block_i->rsv_window_node;
if (!rsv_is_empty(&rsv->rsv_window)) {
spin_lock(rsv_lock);
- rsv_window_remove(inode->i_sb, rsv);
+ if (!rsv_is_empty(&rsv->rsv_window))
+ rsv_window_remove(inode->i_sb, rsv);
spin_unlock(rsv_lock);
}
}
unsigned long bit;
unsigned long i;
unsigned long overflow;
- struct ext3_group_desc * gdp;
+ struct ext3_group_desc * desc;
struct ext3_super_block * es;
struct ext3_sb_info *sbi;
int err = 0, ret;
+ unsigned group_freed;
*pdquot_freed_blocks = 0;
sbi = EXT3_SB(sb);
- es = EXT3_SB(sb)->s_es;
+ es = sbi->s_es;
if (block < le32_to_cpu(es->s_first_data_block) ||
block + count < block ||
block + count > le32_to_cpu(es->s_blocks_count)) {
goto error_return;
}
- ext3_debug ("freeing block %lu\n", block);
+ ext3_debug ("freeing block(s) %lu-%lu\n", block, block + count - 1);
do_more:
overflow = 0;
bitmap_bh = read_block_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
- gdp = ext3_get_group_desc (sb, block_group, &gd_bh);
- if (!gdp)
+ desc = ext3_get_group_desc (sb, block_group, &gd_bh);
+ if (!desc)
goto error_return;
- if (in_range (le32_to_cpu(gdp->bg_block_bitmap), block, count) ||
- in_range (le32_to_cpu(gdp->bg_inode_bitmap), block, count) ||
- in_range (block, le32_to_cpu(gdp->bg_inode_table),
- EXT3_SB(sb)->s_itb_per_group) ||
- in_range (block + count - 1, le32_to_cpu(gdp->bg_inode_table),
- EXT3_SB(sb)->s_itb_per_group))
+ if (in_range (le32_to_cpu(desc->bg_block_bitmap), block, count) ||
+ in_range (le32_to_cpu(desc->bg_inode_bitmap), block, count) ||
+ in_range (block, le32_to_cpu(desc->bg_inode_table),
+ sbi->s_itb_per_group) ||
+ in_range (block + count - 1, le32_to_cpu(desc->bg_inode_table),
+ sbi->s_itb_per_group))
ext3_error (sb, "ext3_free_blocks",
"Freeing blocks in system zones - "
"Block = %lu, count = %lu",
*/
/* @@@ check errors */
BUFFER_TRACE(bitmap_bh, "getting undo access");
- err = ext3_journal_get_undo_access(handle, bitmap_bh, NULL);
+ err = ext3_journal_get_undo_access(handle, bitmap_bh);
if (err)
goto error_return;
jbd_lock_bh_state(bitmap_bh);
- for (i = 0; i < count; i++) {
+ for (i = 0, group_freed = 0; i < count; i++) {
/*
* An HJ special. This is expensive...
*/
}
jbd_lock_bh_state(bitmap_bh);
#endif
+ if (need_resched()) {
+ jbd_unlock_bh_state(bitmap_bh);
+ cond_resched();
+ jbd_lock_bh_state(bitmap_bh);
+ }
/* @@@ This prevents newly-allocated data from being
* freed and then reallocated within the same
* transaction.
jbd_lock_bh_state(bitmap_bh);
BUFFER_TRACE(bitmap_bh, "bit already cleared");
} else {
- (*pdquot_freed_blocks)++;
+ group_freed++;
}
}
jbd_unlock_bh_state(bitmap_bh);
spin_lock(sb_bgl_lock(sbi, block_group));
- gdp->bg_free_blocks_count =
- cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) +
- *pdquot_freed_blocks);
+ desc->bg_free_blocks_count =
+ cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) +
+ group_freed);
spin_unlock(sb_bgl_lock(sbi, block_group));
percpu_counter_mod(&sbi->s_freeblocks_counter, count);
BUFFER_TRACE(gd_bh, "dirtied group descriptor block");
ret = ext3_journal_dirty_metadata(handle, gd_bh);
if (!err) err = ret;
+ *pdquot_freed_blocks += group_freed;
if (overflow && !err) {
block += count;
}
ext3_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks);
if (dquot_freed_blocks) {
- DLIMIT_FREE_BLOCK(sb, inode->i_xid, dquot_freed_blocks);
+ DLIMIT_FREE_BLOCK(inode, dquot_freed_blocks);
DQUOT_FREE_BLOCK(inode, dquot_freed_blocks);
}
return;
*/
static int
ext3_try_to_allocate(struct super_block *sb, handle_t *handle, int group,
- struct buffer_head *bitmap_bh, int goal, struct reserve_window *my_rsv)
+ struct buffer_head *bitmap_bh, int goal, struct ext3_reserve_window *my_rsv)
{
int group_first_block, start, end;
* group. The search will end when we found the start of next
* possible reservable space is out of this boundary.
* This could handle the cross boundary reservation window
- * request.
+ * request.
*
* basically we search from the given range, rather than the whole
* reservation double linked list, (start_block, last_block)
* to find a free region that is of my size and has not
* been reserved.
*
- * on succeed, it returns the reservation window to be appended to.
- * failed, return NULL.
*/
-static struct reserve_window_node *find_next_reservable_window(
- struct reserve_window_node *search_head,
- unsigned long size, int *start_block,
+static int find_next_reservable_window(
+ struct ext3_reserve_window_node *search_head,
+ struct ext3_reserve_window_node *my_rsv,
+ struct super_block * sb, int start_block,
int last_block)
{
struct rb_node *next;
- struct reserve_window_node *rsv, *prev;
+ struct ext3_reserve_window_node *rsv, *prev;
int cur;
+ int size = my_rsv->rsv_goal_size;
/* TODO: make the start of the reservation window byte-aligned */
/* cur = *start_block & ~7;*/
- cur = *start_block;
+ cur = start_block;
rsv = search_head;
- if (!rsv)
- return NULL;
-
+ if (!rsv)
+ return -1;
+
while (1) {
if (cur <= rsv->rsv_end)
cur = rsv->rsv_end + 1;
* space with expected-size (or more)...
*/
if (cur > last_block)
- return NULL; /* fail */
+ return -1; /* fail */
prev = rsv;
next = rb_next(&rsv->rsv_node);
- rsv = list_entry(next, struct reserve_window_node, rsv_node);
+ rsv = list_entry(next,struct ext3_reserve_window_node,rsv_node);
- /*
- * Reached the last reservation, we can just append to the
- * previous one.
+ /*
+ * Reached the last reservation, we can just append to the
+ * previous one.
*/
if (!next)
break;
* return the reservation window that we could append to.
* succeed.
*/
- *start_block = cur;
- return prev;
+
+ if ((prev != my_rsv) && (!rsv_is_empty(&my_rsv->rsv_window)))
+ rsv_window_remove(sb, my_rsv);
+
+ /*
+ * Let's book the whole avaliable window for now. We will check the
+ * disk bitmap later and then, if there are free blocks then we adjust
+ * the window size if it's larger than requested.
+ * Otherwise, we will remove this node from the tree next time
+ * call find_next_reservable_window.
+ */
+ my_rsv->rsv_start = cur;
+ my_rsv->rsv_end = cur + size - 1;
+ my_rsv->rsv_alloc_hit = 0;
+
+ if (prev != my_rsv)
+ ext3_rsv_window_add(sb, my_rsv);
+
+ return 0;
}
/**
* alloc_new_reservation()--allocate a new reservation window
- * if there is an existing reservation, discard it first
- * then allocate the new one from there
- * otherwise allocate the new reservation from the given
- * start block, or the beginning of the group, if a goal
- * is not given.
*
* To make a new reservation, we search part of the filesystem
- * reservation list (the list that inside the group).
- *
- * If we have a old reservation, the search goal is the end of
- * last reservation. If we do not have a old reservation, then we
- * start from a given goal, or the first block of the group, if
- * the goal is not given.
+ * reservation list (the list that inside the group). We try to
+ * allocate a new reservation window near the allocation goal,
+ * or the beginning of the group, if there is no goal.
*
* We first find a reservable space after the goal, then from
* there, we check the bitmap for the first free block after
*
* @goal: The goal (group-relative). It is where the search for a
* free reservable space should start from.
- * if we have a old reservation, start_block is the end of
- * old reservation. Otherwise,
* if we have a goal(goal >0 ), then start from there,
* no goal(goal = -1), we start from the first block
* of the group.
* @sb: the super block
* @group: the group we are trying to allocate in
* @bitmap_bh: the block group block bitmap
+ *
*/
-static int alloc_new_reservation(struct reserve_window_node *my_rsv,
+static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv,
int goal, struct super_block *sb,
unsigned int group, struct buffer_head *bitmap_bh)
{
- struct reserve_window_node *search_head;
+ struct ext3_reserve_window_node *search_head;
int group_first_block, group_end_block, start_block;
int first_free_block;
- int reservable_space_start;
- struct reserve_window_node *prev_rsv;
struct rb_root *fs_rsv_root = &EXT3_SB(sb)->s_rsv_window_root;
unsigned long size;
+ int ret;
+ spinlock_t *rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock;
group_first_block = le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block) +
group * EXT3_BLOCKS_PER_GROUP(sb);
else
start_block = goal + group_first_block;
- size = atomic_read(&my_rsv->rsv_goal_size);
- /* if we have a old reservation, start the search from the old rsv */
+ size = my_rsv->rsv_goal_size;
+
if (!rsv_is_empty(&my_rsv->rsv_window)) {
/*
* if the old reservation is cross group boundary
+ * and if the goal is inside the old reservation window,
* we will come here when we just failed to allocate from
* the first part of the window. We still have another part
* that belongs to the next group. In this case, there is no
*/
if ((my_rsv->rsv_start <= group_end_block) &&
- (my_rsv->rsv_end > group_end_block))
+ (my_rsv->rsv_end > group_end_block) &&
+ (start_block >= my_rsv->rsv_start))
return -1;
- /* remember where we are before we discard the old one */
- if (my_rsv->rsv_end + 1 > start_block)
- start_block = my_rsv->rsv_end + 1;
- search_head = my_rsv;
- if ((atomic_read(&my_rsv->rsv_alloc_hit) >
+ if ((my_rsv->rsv_alloc_hit >
(my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) {
/*
* if we previously allocation hit ration is greater than half
size = size * 2;
if (size > EXT3_MAX_RESERVE_BLOCKS)
size = EXT3_MAX_RESERVE_BLOCKS;
- atomic_set(&my_rsv->rsv_goal_size, size);
+ my_rsv->rsv_goal_size= size;
}
}
- else {
- /*
- * we don't have a reservation,
- * we set our goal(start_block) and
- * the list head for the search
- */
- search_head = search_reserve_window(fs_rsv_root, start_block);
- }
+
+ spin_lock(rsv_lock);
+ /*
+ * shift the search start to the window near the goal block
+ */
+ search_head = search_reserve_window(fs_rsv_root, start_block);
/*
* find_next_reservable_window() simply finds a reservable window
* need to check the bitmap after we found a reservable window.
*/
retry:
- prev_rsv = find_next_reservable_window(search_head, size,
- &start_block, group_end_block);
- if (prev_rsv == NULL)
- goto failed;
- reservable_space_start = start_block;
+ ret = find_next_reservable_window(search_head, my_rsv, sb,
+ start_block, group_end_block);
+
+ if (ret == -1) {
+ if (!rsv_is_empty(&my_rsv->rsv_window))
+ rsv_window_remove(sb, my_rsv);
+ spin_unlock(rsv_lock);
+ return -1;
+ }
+
/*
* On success, find_next_reservable_window() returns the
* reservation window where there is a reservable space after it.
* block. Search start from the start block of the reservable space
* we just found.
*/
+ spin_unlock(rsv_lock);
first_free_block = bitmap_search_next_usable_block(
- reservable_space_start - group_first_block,
+ my_rsv->rsv_start - group_first_block,
bitmap_bh, group_end_block - group_first_block + 1);
if (first_free_block < 0) {
* no free block left on the bitmap, no point
* to reserve the space. return failed.
*/
- goto failed;
+ spin_lock(rsv_lock);
+ if (!rsv_is_empty(&my_rsv->rsv_window))
+ rsv_window_remove(sb, my_rsv);
+ spin_unlock(rsv_lock);
+ return -1; /* failed */
}
+
start_block = first_free_block + group_first_block;
/*
* check if the first free block is within the
- * free space we just found
+ * free space we just reserved
*/
- if ((start_block >= reservable_space_start) &&
- (start_block < reservable_space_start + size))
- goto found_rsv_window;
+ if (start_block >= my_rsv->rsv_start && start_block < my_rsv->rsv_end)
+ return 0; /* success */
/*
* if the first free bit we found is out of the reservable space
- * this means there is no free block on the reservable space
- * we should continue search for next reservable space,
+ * continue search for next reservable space,
* start from where the free block is,
* we also shift the list head to where we stopped last time
*/
- search_head = prev_rsv;
+ search_head = my_rsv;
+ spin_lock(rsv_lock);
goto retry;
-
-found_rsv_window:
- /*
- * great! the reservable space contains some free blocks.
- * if the search returns that we should add the new
- * window just next to where the old window, we don't
- * need to remove the old window first then add it to the
- * same place, just update the new start and new end.
- */
- if (my_rsv != prev_rsv) {
- if (!rsv_is_empty(&my_rsv->rsv_window))
- rsv_window_remove(sb, my_rsv);
- }
- my_rsv->rsv_start = reservable_space_start;
- my_rsv->rsv_end = my_rsv->rsv_start + size - 1;
- atomic_set(&my_rsv->rsv_alloc_hit, 0);
- if (my_rsv != prev_rsv) {
- rsv_window_add(sb, my_rsv);
- }
- return 0; /* succeed */
-failed:
- return -1; /* failed */
}
/*
* allocation within the reservation window.
*
* This will avoid keeping on searching the reservation list again and
- * again when someboday is looking for a free block (without
+ * again when somebody is looking for a free block (without
* reservation), and there are lots of free blocks, but they are all
* being reserved.
*
static int
ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
unsigned int group, struct buffer_head *bitmap_bh,
- int goal, struct reserve_window_node * my_rsv,
+ int goal, struct ext3_reserve_window_node * my_rsv,
int *errp)
{
- spinlock_t *rsv_lock;
unsigned long group_first_block;
int ret = 0;
int fatal;
- int credits = 0;
*errp = 0;
* if the buffer is in BJ_Forget state in the committing transaction.
*/
BUFFER_TRACE(bitmap_bh, "get undo access for new block");
- fatal = ext3_journal_get_undo_access(handle, bitmap_bh, &credits);
+ fatal = ext3_journal_get_undo_access(handle, bitmap_bh);
if (fatal) {
*errp = fatal;
return -1;
ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, goal, NULL);
goto out;
}
- rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock;
/*
* goal is a group relative block number (if there is a goal)
* 0 < goal < EXT3_BLOCKS_PER_GROUP(sb)
* then we could go to allocate from the reservation window directly.
*/
while (1) {
- struct reserve_window rsv_copy;
- unsigned int seq;
-
- do {
- seq = read_seqbegin(&my_rsv->rsv_seqlock);
- rsv_copy._rsv_start = my_rsv->rsv_start;
- rsv_copy._rsv_end = my_rsv->rsv_end;
- } while (read_seqretry(&my_rsv->rsv_seqlock, seq));
-
- if (rsv_is_empty(&rsv_copy) || (ret < 0) ||
- !goal_in_my_reservation(&rsv_copy, goal, group, sb)) {
- spin_lock(rsv_lock);
- write_seqlock(&my_rsv->rsv_seqlock);
+ if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) ||
+ !goal_in_my_reservation(&my_rsv->rsv_window, goal, group, sb)) {
ret = alloc_new_reservation(my_rsv, goal, sb,
group, bitmap_bh);
- rsv_copy._rsv_start = my_rsv->rsv_start;
- rsv_copy._rsv_end = my_rsv->rsv_end;
- write_sequnlock(&my_rsv->rsv_seqlock);
- spin_unlock(rsv_lock);
if (ret < 0)
break; /* failed */
- if (!goal_in_my_reservation(&rsv_copy, goal, group, sb))
+ if (!goal_in_my_reservation(&my_rsv->rsv_window, goal, group, sb))
goal = -1;
}
- if ((rsv_copy._rsv_start >= group_first_block + EXT3_BLOCKS_PER_GROUP(sb))
- || (rsv_copy._rsv_end < group_first_block))
+ if ((my_rsv->rsv_start >= group_first_block + EXT3_BLOCKS_PER_GROUP(sb))
+ || (my_rsv->rsv_end < group_first_block))
BUG();
ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, goal,
- &rsv_copy);
+ &my_rsv->rsv_window);
if (ret >= 0) {
- if (!read_seqretry(&my_rsv->rsv_seqlock, seq))
- atomic_inc(&my_rsv->rsv_alloc_hit);
+ my_rsv->rsv_alloc_hit++;
break; /* succeed */
}
}
}
BUFFER_TRACE(bitmap_bh, "journal_release_buffer");
- ext3_journal_release_buffer(handle, bitmap_bh, credits);
+ ext3_journal_release_buffer(handle, bitmap_bh);
return ret;
}
struct ext3_group_desc *gdp;
struct ext3_super_block *es;
struct ext3_sb_info *sbi;
- struct reserve_window_node *my_rsv = NULL;
+ struct ext3_reserve_window_node *my_rsv = NULL;
+ struct ext3_block_alloc_info *block_i;
+ unsigned short windowsz = 0;
#ifdef EXT3FS_DEBUG
static int goal_hits, goal_attempts;
#endif
*errp = -EDQUOT;
return 0;
}
- if (DLIMIT_ALLOC_BLOCK(sb, inode->i_xid, 1))
- goto out_dlimit;
+ if (DLIMIT_ALLOC_BLOCK(inode, 1))
+ goto out_dlimit;
sbi = EXT3_SB(sb);
es = EXT3_SB(sb)->s_es;
ext3_debug("goal=%lu.\n", goal);
- if (test_opt(sb, RESERVATION) && S_ISREG(inode->i_mode))
- my_rsv = &EXT3_I(inode)->i_rsv_window;
-#warning MEF was if (!ext3_has_free_blocks(sbi)) in 1.11-FC2
+ /*
+ * Allocate a block from reservation only when
+ * filesystem is mounted with reservation(default,-o reservation), and
+ * it's a regular file, and
+ * the desired window size is greater than 0 (One could use ioctl
+ * command EXT3_IOC_SETRSVSZ to set the window size to 0 to turn off
+ * reservation on that particular file)
+ */
+ block_i = EXT3_I(inode)->i_block_alloc_info;
+ if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0))
+ my_rsv = &block_i->rsv_window_node;
+
if (!ext3_has_free_blocks(sb)) {
*errp = -ENOSPC;
goto out;
goal_group = group_no;
retry:
free_blocks = le16_to_cpu(gdp->bg_free_blocks_count);
+ /*
+ * if there is not enough free blocks to make a new resevation
+ * turn off reservation for this allocation
+ */
+ if (my_rsv && (free_blocks < windowsz)
+ && (rsv_is_empty(&my_rsv->rsv_window)))
+ my_rsv = NULL;
+
if (free_blocks > 0) {
ret_block = ((goal - le32_to_cpu(es->s_first_data_block)) %
EXT3_BLOCKS_PER_GROUP(sb));
goto out;
}
free_blocks = le16_to_cpu(gdp->bg_free_blocks_count);
- if (free_blocks <= 0)
+ /*
+ * skip this group if the number of
+ * free blocks is less than half of the reservation
+ * window size.
+ */
+ if (free_blocks <= (windowsz/2))
continue;
brelse(bitmap_bh);
*errp = -EIO;
out:
if (!performed_allocation)
- DLIMIT_FREE_BLOCK(sb, inode->i_xid, 1);
+ DLIMIT_FREE_BLOCK(inode, 1);
out_dlimit:
if (fatal) {
*errp = fatal;
unsigned long desc_count;
struct ext3_group_desc *gdp;
int i;
- unsigned long ngroups;
+ unsigned long ngroups = EXT3_SB(sb)->s_groups_count;
#ifdef EXT3FS_DEBUG
struct ext3_super_block *es;
unsigned long bitmap_count, x;
struct buffer_head *bitmap_bh = NULL;
- lock_super(sb);
es = EXT3_SB(sb)->s_es;
desc_count = 0;
bitmap_count = 0;
gdp = NULL;
- for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) {
+
+ smp_rmb();
+ for (i = 0; i < ngroups; i++) {
gdp = ext3_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
brelse(bitmap_bh);
printk("ext3_count_free_blocks: stored = %u, computed = %lu, %lu\n",
le32_to_cpu(es->s_free_blocks_count), desc_count, bitmap_count);
- unlock_super(sb);
return bitmap_count;
#else
desc_count = 0;
- ngroups = EXT3_SB(sb)->s_groups_count;
smp_rmb();
for (i = 0; i < ngroups; i++) {
gdp = ext3_get_group_desc(sb, i, NULL);
#endif
}
-static inline int block_in_use(unsigned long block,
- struct super_block * sb,
- unsigned char * map)
+static inline int
+block_in_use(unsigned long block, struct super_block *sb, unsigned char *map)
{
return ext3_test_bit ((block -
le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block)) %
static inline int test_root(int a, int b)
{
- if (a == 0)
- return 1;
- while (1) {
- if (a == 1)
- return 1;
- if (a % b)
- return 0;
- a = a / b;
- }
+ int num = b;
+
+ while (a > num)
+ num *= b;
+ return num == a;
}
-int ext3_group_sparse(int group)
+static int ext3_group_sparse(int group)
{
- return (test_root(group, 3) || test_root(group, 5) ||
- test_root(group, 7));
+ if (group <= 1)
+ return 1;
+ if (!(group & 1))
+ return 0;
+ return (test_root(group, 7) || test_root(group, 5) ||
+ test_root(group, 3));
}
/**
return EXT3_SB(sb)->s_gdb_count;
}
-#ifdef CONFIG_EXT3_CHECK
-/* Called at mount-time, super-block is locked */
-void ext3_check_blocks_bitmap (struct super_block * sb)
-{
- struct ext3_super_block *es;
- unsigned long desc_count, bitmap_count, x, j;
- unsigned long desc_blocks;
- struct buffer_head *bitmap_bh = NULL;
- struct ext3_group_desc *gdp;
- int i;
-
- es = EXT3_SB(sb)->s_es;
- desc_count = 0;
- bitmap_count = 0;
- gdp = NULL;
- for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) {
- gdp = ext3_get_group_desc (sb, i, NULL);
- if (!gdp)
- continue;
- desc_count += le16_to_cpu(gdp->bg_free_blocks_count);
- brelse(bitmap_bh);
- bitmap_bh = read_block_bitmap(sb, i);
- if (bitmap_bh == NULL)
- continue;
-
- if (ext3_bg_has_super(sb, i) &&
- !ext3_test_bit(0, bitmap_bh->b_data))
- ext3_error(sb, __FUNCTION__,
- "Superblock in group %d is marked free", i);
-
- desc_blocks = ext3_bg_num_gdb(sb, i);
- for (j = 0; j < desc_blocks; j++)
- if (!ext3_test_bit(j + 1, bitmap_bh->b_data))
- ext3_error(sb, __FUNCTION__,
- "Descriptor block #%ld in group "
- "%d is marked free", j, i);
-
- if (!block_in_use (le32_to_cpu(gdp->bg_block_bitmap),
- sb, bitmap_bh->b_data))
- ext3_error (sb, "ext3_check_blocks_bitmap",
- "Block bitmap for group %d is marked free",
- i);
-
- if (!block_in_use (le32_to_cpu(gdp->bg_inode_bitmap),
- sb, bitmap_bh->b_data))
- ext3_error (sb, "ext3_check_blocks_bitmap",
- "Inode bitmap for group %d is marked free",
- i);
-
- for (j = 0; j < EXT3_SB(sb)->s_itb_per_group; j++)
- if (!block_in_use (le32_to_cpu(gdp->bg_inode_table) + j,
- sb, bitmap_bh->b_data))
- ext3_error (sb, "ext3_check_blocks_bitmap",
- "Block #%d of the inode table in "
- "group %d is marked free", j, i);
-
- x = ext3_count_free(bitmap_bh, sb->s_blocksize);
- if (le16_to_cpu(gdp->bg_free_blocks_count) != x)
- ext3_error (sb, "ext3_check_blocks_bitmap",
- "Wrong free blocks count for group %d, "
- "stored = %d, counted = %lu", i,
- le16_to_cpu(gdp->bg_free_blocks_count), x);
- bitmap_count += x;
- }
- brelse(bitmap_bh);
- if (le32_to_cpu(es->s_free_blocks_count) != bitmap_count)
- ext3_error (sb, "ext3_check_blocks_bitmap",
- "Wrong free blocks count in super block, "
- "stored = %lu, counted = %lu",
- (unsigned long)le32_to_cpu(es->s_free_blocks_count),
- bitmap_count);
-}
-#endif